Optimum Production Run Length and Process Mean Setting

Similar documents
International Journal of Advances in Science and Technology (IJAST)

School of Economics and Management, Tongji University, Shanghai , China. Correspondence should be addressed to Bing-Bing QIU,

EXPONENTIAL DEPENDENT DEMAND RATE ECONOMIC PRODUCTION QUANTITY (EPQ) MODEL WITH FOR REDUCTION DELIVERY POLICY

OPTIMAL POLICY FOR A SIMPLE SUPPLY CHAIN SYSTEM WITH DEFECTIVE ITEMS AND RETURNED COST UNDER SCREENING ERRORS

A Simple Utility Approach to Private Equity Sales

Economic Production Quantity (EPQ) Model with Time- Dependent Demand and Reduction Delivery Policy

An Integrated Production Inventory System for. Perishable Items with Fixed and Linear Backorders

Keywords: Single-vendor Inventory Control System, Potential Demand, Machine Failure, Participation in the Chain, Heuristic Algorithm

A Detailed Price Discrimination Example

EPQ MODEL FOR IMPERFECT PRODUCTION PROCESSES WITH REWORK AND RANDOM PREVENTIVE MACHINE TIME FOR DETERIORATING ITEMS AND TRENDED DEMAND

SUPPLIER SELECTION IN A CLOSED-LOOP SUPPLY CHAIN NETWORK

Math 120 Final Exam Practice Problems, Form: A

A Profit-Maximizing Production Lot sizing Decision Model with Stochastic Demand

Package SCperf. February 19, 2015

Supply Chain Planning Considering the Production of Defective Products

An Entropic Order Quantity (EnOQ) Model. with Post Deterioration Cash Discounts

STATISTICAL QUALITY CONTROL (SQC)

Research Article An Optimal Returned Policy for a Reverse Logistics Inventory Model with Backorders

Lecture 2. Marginal Functions, Average Functions, Elasticity, the Marginal Principle, and Constrained Optimization

TAGUCHI APPROACH TO DESIGN OPTIMIZATION FOR QUALITY AND COST: AN OVERVIEW. Resit Unal. Edwin B. Dean

A QUEUEING-INVENTORY SYSTEM WITH DEFECTIVE ITEMS AND POISSON DEMAND.

ECON 459 Game Theory. Lecture Notes Auctions. Luca Anderlini Spring 2015

Lecture 3. Linear Programming. 3B1B Optimization Michaelmas 2015 A. Zisserman. Extreme solutions. Simplex method. Interior point method

OPTIMAL CONTROL OF A PRODUCTION INVENTORY SYSTEM WITH GENERALIZED EXPONENTIAL DISTRIBUTED DETERIORATION

Fuzzy regression model with fuzzy input and output data for manpower forecasting

PROPERTIES OF THE SAMPLE CORRELATION OF THE BIVARIATE LOGNORMAL DISTRIBUTION

A simulation study on supply chain performance with uncertainty using contract. Creative Commons: Attribution 3.0 Hong Kong License

BINOMIAL OPTIONS PRICING MODEL. Mark Ioffe. Abstract

A NEW APPROACH FOR MEASUREMENT OF THE EFFICIENCY OF C pm AND C pmk CONTROL CHARTS

Equations, Inequalities & Partial Fractions

Optimization of Preventive Maintenance Scheduling in Processing Plants

Overview. Essential Questions. Precalculus, Quarter 4, Unit 4.5 Build Arithmetic and Geometric Sequences and Series

Product Differentiation In homogeneous goods markets, price competition leads to perfectly competitive outcome, even with two firms Price competition

INTEGRATED OPTIMIZATION OF SAFETY STOCK

Nonparametric adaptive age replacement with a one-cycle criterion

Winkler, D. T. "The Cost of Trade Credit: A Net Present Value Perspective." Journal of Business and Economic Studies, vol. 3, no. 1, 1996, pp

Lecture 3: Linear methods for classification

Sensitivity Analysis 3.1 AN EXAMPLE FOR ANALYSIS

Betting rules and information theory

How To Use Statgraphics Centurion Xvii (Version 17) On A Computer Or A Computer (For Free)

Tolerance Charts. Dr. Pulak M. Pandey.

1 Calculus of Several Variables

Inventory Management - A Teaching Note

Epipolar Geometry. Readings: See Sections 10.1 and 15.6 of Forsyth and Ponce. Right Image. Left Image. e(p ) Epipolar Lines. e(q ) q R.

3.3. Solving Polynomial Equations. Introduction. Prerequisites. Learning Outcomes

A Production Planning Problem

FUZZY APPROACH ON OPTIMAL ORDERING STRATEGY IN INVENTORY AND PRICING MODEL WITH DETERIORATING ITEMS

Investment Statistics: Definitions & Formulas

Analysis of a Production/Inventory System with Multiple Retailers

Discrete Optimization

An integrated Single Vendor-Single Buyer Production Inventory System Incorporating Warehouse Sizing Decisions 창고 크기의사결정을 포함한 단일 공급자구매자 생산재고 통합관리 시스템

Second degree price discrimination

Journal of Emerging Trends in Engineering and Applied Sciences (JETEAS) 4(1): (ISSN: )

Multi-variable Calculus and Optimization

FEGYVERNEKI SÁNDOR, PROBABILITY THEORY AND MATHEmATICAL

2013 MBA Jump Start Program

1 The EOQ and Extensions

Introduction to General and Generalized Linear Models

On the Efficiency of Competitive Stock Markets Where Traders Have Diverse Information

Research Article Two-Period Inventory Control with Manufacturing and Remanufacturing under Return Compensation Policy

Questions and Answers

STRATEGIC CAPACITY PLANNING USING STOCK CONTROL MODEL

An integrated just-in-time inventory system with stock-dependent demand

Constrained Bayes and Empirical Bayes Estimator Applications in Insurance Pricing

Understanding the Impact of Weights Constraints in Portfolio Theory

Least Squares Estimation

1 Math 1313 Final Review Final Review for Finite. 1. Find the equation of the line containing the points 1, 2)

Numerical Methods for Option Pricing

Confidence Intervals for One Standard Deviation Using Standard Deviation

CURVE FITTING LEAST SQUARES APPROXIMATION

Logistics Management Customer Service. Özgür Kabak, Ph.D.

Grade Level Year Total Points Core Points % At Standard %

Enhancing the SNR of the Fiber Optic Rotation Sensor using the LMS Algorithm

A Fast Algorithm for Multilevel Thresholding

Optimal Hedging of Uncertain Foreign Currency Returns. 12/12/2011 SLJ Macro Partners Fatih Yilmaz

A Stochastic Inventory Placement Model for a Multi-echelon Seasonal Product Supply Chain with Multiple Retailers

A Genetic Algorithm Approach for Solving a Flexible Job Shop Scheduling Problem

Revenue Management for Transportation Problems

A logistic approximation to the cumulative normal distribution

Testing against a Change from Short to Long Memory

Chapter 3 RANDOM VARIATE GENERATION

Modern Optimization Methods for Big Data Problems MATH11146 The University of Edinburgh

1 Mathematical Models of Cost, Revenue and Profit

Chapter 11 Monte Carlo Simulation

Accounting 402 Illustration of a change in inventory method

Testing against a Change from Short to Long Memory

The normal approximation to the binomial

Mechanics 1: Conservation of Energy and Momentum

ISyE 2030 Test 2 Solutions

Chapter 21: The Discounted Utility Model

SOLVING LINEAR SYSTEMS

ECO 199 B GAMES OF STRATEGY Spring Term 2004 PROBLEM SET 4 B DRAFT ANSWER KEY

INVENTORY MODELS WITH STOCK- AND PRICE- DEPENDENT DEMAND FOR DETERIORATING ITEMS BASED ON LIMITED SHELF SPACE

Chapter 7. Production, Capacity and Material Planning

MAT12X Intermediate Algebra

4 The M/M/1 queue. 4.1 Time-dependent behaviour

Lesson 20. Probability and Cumulative Distribution Functions

6.207/14.15: Networks Lecture 15: Repeated Games and Cooperation

Largest Fixed-Aspect, Axis-Aligned Rectangle

Math 370/408, Spring 2008 Prof. A.J. Hildebrand. Actuarial Exam Practice Problem Set 3 Solutions

Transcription:

Tamkang Journal of Science and Engineering, Vol. 13, No., pp. 157 164 (010) 157 Optimum Production Run Length and Process Mean Setting Chung-Ho Chen Institute of Industrial Management, Southern Taiwan University, Yungkang, Taiwan 710, R.O.C. Abstract In this paper, the author presented a modified Chen and Chung s model considering that the process mean may not be equal to the target value when the process is in the in-control state. Taguchi s asymmetric quadratic quality loss function will be applied in evaluating the product quality. Subsequently, the modified economic manufacturing quantity (EMQ) model based on modified the Chen and Chung s model is formulated for obtaining the expected total cost per year. The numerical result shows that for an in-control process, the modified EMQ model when the process mean is not equal to the target value has smaller expected total cost per year than that of the modified EMQ model when the process mean is equal to the target value. Key Words: Economic Manufacturing Quantity (EMQ), Production Run Length, Process Mean, Taguchi s Asymmetric Quadratic Quality Loss Function 1. Introduction *Corresponding author. E-mail: chench@mail.stut.edu.tw The traditional economic manufacturing quantity (EMQ) model assumes that the product for a manufacturing process is perfect. Hence, the occurrence of a defective product for this model has been neglected. Previous researchers, like Porteus [1] and Rosenblatt and Lee [,3], first proposed the imperfect quality of the EMQ model. Subsequently, some works addressed the integrated model on production, inspection, maintenance, and quality. Lee and Rosenblatt [4,5] and Lee and Park [6] introduced some inspection and maintenance mechanisms to monitor a production process. Rahim [7] and Rahim and Ben-Daya [8] presented an integrated model for jointly determining the production quantity, inspection policy, and parameters of a statistical control chart. Al-Sultan [9] pointed out that optimization methods have been successfully applied to the economic models integrating production, maintenance, and quality. Wright and Mehrez [10] presented a literature review that includes the connection between quality and inventory. Recently, Tahera et al. [11] addressed a review on a joint determination of process level and production run length. In 1997, Sana et al. [1,13] proposed a volume flexible inventory model considering the demand rate of defective items sold at a reduced price as a function of the reduction rate of the selling price. They further addressed the imperfect production system including the unit production cost taken to be a function of the finite production rate. In 009, Sana [14,15] considered the production-inventory model with imperfect production system for defective items restored to their original quality by rework. His model also discusses and provides an optimal solution for product reliability parameter and dynamic production rate. Economic selection of process parameters has been an important topic in modern statistical process control. The optimum process mean setting has an effect on the expected total profit/cost, defective fraction, inspection/ reprocessing cost. Recently, there is also considerable attention paid to the filling/canning problem for determining the optimum manufacturing target and other important parameters. Jang et al. [16] proposed the optimum process mean and production cycle settings for a linear

158 Chung-Ho Chen trend process. Young et al. [17] discussed the process mean setting under the variance reduction and a specified process capability. Lee and Elsayed [18] presented a two-stage screening method for determining the process mean and screening limits. Anis [19] considered the cost of non-conformance for obtaining an optimum process mean. Hong and Cho [0] proposed the problem of a joint design of process mean and tolerance limits based on measurement error. In 1986, Taguchi [1] presented the quadratic quality loss function for redefining the product quality. According to him, product quality is the society s loss when the product is sold to the customer. The total loss of a society includes the producer s loss (the set-up cost, the holding cost, the inspection cost, and the production cost for the producer) and the consumer s loss (including the quality cost). An optimum product quality from Taguchi s [1] point should be defined as a product quality characteristic with minimum bias and variance. Recently, Taguchi s [1] quality loss function was successfully applied in the problem of optimum process mean settings. Kim and Cho [] adopted a truncated Weibull quality characteristic and quadratic quality loss function for determining an optimum process mean. Rahim and Tuffaha [3] proposed the quadratic and linear quality loss functions applied in the determination of production run length and process mean. Chan and Ibrahim [4] and Chan et al. [5,6] addressed the multivariate quadratic quality loss functions applied in the determination of optimum process mean for the nominal-the-best, smaller-the-better, and larger-thebetter quality characteristics, respectively. Teeravaraprug [7] adopted the quadratic quality loss function for evaluating the quality cost of a product for two different markets and obtained the optimum process mean based on maximizing the expected profit per item. Chen [8,9] proposed a modified Pulak and Al-Sultan s [30] model with quality loss for determining the optimum process parameters under the rectifying inspection plan. In 1996, Chen and Chung [31] presented the quality selection problem in imperfect production systems for obtaining the optimum production run length and target level. Rahim and Tuffaha [3] further proposed the modified Chen and Chung s [31] model with quality loss and sampling inspection. However, the in-control process mean is assumed to be equal to the target value and no defective item exists when the production process is in statistical control for the Chen and Chung s [31] and Rahim and Tuffaha s [3] models. In fact, the process mean may not be equal to the target value and a defective item may exists when the production process is in the in-control state. In this paper, the author further proposes a modified EMQ model based on the modified Chen and Chung s [31] model considering that the process mean may not be equal to the target value when the process is in the in-control state. Taguchi s [1] asymmetric quadratic quality loss function will be applied in evaluating a product s quality. By solving the modified EMQ model, one can obtain the optimum production run length and process mean with the minimum expected total cost per year. Finally, the sensitivity analyses of cost parameters will be provided for illustration.. Taguchi s Asymmetric Quadratic Quality Loss Function Assume that y is the nominal-is-best quality characteristic. Then the asymmetric quality loss function is defined as (1) When the loss function is k (y m), the constant k can be obtained as follows: Find the limit value, above which an item performs unsatisfactorily for example, in more than 50% of cases and assess the corresponding consumer s penalty A. Substituting A and in the loss function for an individual item, i.e., L(y) =k (y m) A, and solving for k,weobtaink. Similarly, we can obtain the constant k 1 as follows: Find the limit value 1, above which an item performs unsatisfactorily for example, in more than 50% of cases and assess the corresponding consumer s penalty A 1.Substituting A 1 and 1 in the loss function for an individual item, i.e., L(y) =k 1 (y m), and solving for k 1,weobtain k. A1 1 1 3. Chen and Chung s Model There are some assumptions made in Chen and

Optimum Production Run Length and Process Mean Setting 159 Chung [31], such as (1) When the production cycle starts, the process is in the in-control state. Once a shift has occurred, the process will remain in an out-of-control state until it is discovered by inspection and followed by some restoration work. Otherwise, the out-of-control state will remain until the end of the production run. () denotes the elapse time until the occurrence of an assignable cause assumed to be exponentially distributed with a mean of 1/ ( is the time parameter and is distributed as exponential when the production process is in the in-control state). (3) The process mean can be adjusted and controlled and the shape of the quality characteristic does not change when the process mean is adjusted or changed. (4) If the value of the quality characteristic is equal to or greater than the lower specification limit, the item is sold at a regular price. On the other hand, the item is sold as a scrap at a reduced price. From Chen and Chung [31, p. 054], the expected gross profit per item at time t is () where g 0 is the expected revenue per item for an incontrol process and g 1 is the expected revenue per item for an out-of-control process. Then the expected total revenue in the production run T is (3) Chen and Chung s [31] model, such as (1) The quality characteristic X is normally distributed with an unknown process mean which can be easily adjusted, but the process variance is known. () When the production cycle starts, the process is in the in-control state. Once a shift has occurred, the process will remain in an out-of-control state until it is discovered by inspection and followed by some restoration work. Otherwise, the out-of-control state will remain until the end of the production run. (3) The asymmetric quadratic quality loss function is used in measuring a product s quality. (4) The elapse time until the occurrence of an assignable cause is assumed to be exponentially distributed with a mean of 1/. (5) The process mean of the in-control process is not equal to the target value. From the above assumption, the expected total cost for a production cycle T should include the set-up cost, the holding cost, and the quality loss per item. The modified model with the quality cost per item for each production cycle T is (5) where g 0 is the expected quality cost of an in-control process, g 1 is the expected quality cost of an out-of-control process, where p is the production rate in pieces per hour. Hence, the expected total profit per item for each production cycle T is 4. Modified Chen and Chung s Model (4) There are some assumptions made in the modified If the process mean of the in-control process is equal 1 to the target value, then one has g0 ( k1 k)and g 1 = {(1 + )[k 1 +(k k 1 ) ( )] + (k 1 k ) ( )}.

160 Chung-Ho Chen 5. Modified Economic Manufacturing Quantity Model To minimize the expected total cost per year, partially differentiate Eq. (8) with respect to T and m, and letting the partial derivatives equal to zero, i.e., There are three main assumptions for the traditional EMQ model: (1) A perfect quality occurs in the production process; () There is no shortage of cost; (3) The demand for the produced item is continuous and constant and all demands must be met (production rate > demand rate). Based on the above assumptions, the expected total cost of the traditional EMQ model which includes the set-up cost and the holding cost is as follows: (9) (10) where (6) (11) where ETC is the expected total cost per year; W is the demand quantity in units per year; d is the demand quantity in units per day; S is the set-up cost for each production run; p is the production rate in units per day; B is the holding cost per unit item per year; T is the production run length. One sets the first derivative of ETC equal to zero, and solves for the economic manufacturing quantity, Q (= pt). We have the optimum economic manufacturing WSp quantity and production run length as QE* = B ( p - d) and TE* = WS, respectively. p( p - d ) B Hence, the expected total cost per year for the modified EMQ model based on the modified Chen and Chung s [31] model with expected quality loss is as follows: (1) It is difficult to show that the Hessian matrix of Eq. (8) is positive definite because the expected total cost per year includes the standard normal probability density function f( ) and the cumulative standard normal distribution function F( ). It is very complex in obtaining the second differentiation in g0 and g1, with respect to m of Eqs. (11) and (1). One cannot prove that Eq. (8) has a closed-form solution. From Eq. (9), one has an explicit expression of T in terms of other variables: (7) Assume that lt is sufficiently small, then e-lt = 1 (lt ). By applying this approximation, Eq. (7) can lt + be rewritten as (8) (13) where

Optimum Production Run Length and Process Mean Setting 161 The expected quality cost of the in-control process is less than that of the out-of-control process because the latter has more defective items than the former. From * * g 0 < g 1, we have T 1 T E. The optimal solution in Eq. (8) is valid only if the expected total cost is convex. The second derivative of Eq. (8) with respect to T is ETP WS 0. Hence, the 3 T pt optimal solution exists for the given parameters. For the given parameters, one can adopt the direct search method for obtaining the economic production run length T * and the optimum process mean *. The solution procedure for the above Eq. (8) is as follows: WS * WS Step 1. Compute TE. p() p( p d) B Step. Set the minimum searched value T, T min = 0.01, and let the maximum searched value T, T max =T * E. Step 3. Set the minimum searched value, min = m 4, and let the maximum searched value, max = m +4. Step 4. Let T = T min and = min. Compute ETC 1 (T, )by Eq. (8). Let T = T min + 0.01. Repeat this step until T = T max. Step 5. Let = min + 0.01. Repeat Step 4 until = max. Step 6. Select the maximum expected total cost from the above Steps 1 5 as the best policy. The corresponding parameters of T * and * values having the minimum ETC 1 (T, ) is the optimal solution. The optimal solution of the modified EMQ model depends on the parameters of cost. The influences of them need to be illustrated using a sensitivity analysis. 6. Numerical Example and Sensitivity Analysis 6.1 Numerical Example Suppose that the demand for an item is W = 0,000 items per year and there are 50 working days per year. The production rate is p = 100 items per day. The holding cost is B = 10 per item per year, and the set-up cost is S = 0 per run. The quality characteristic is normally distributed with a known standard deviation =0.05.Let the magnitude of a mean shift = 1, the quality loss coefficients k 1 =1andk =, the target value m =10,and the parameter of the exponential distribution, = 0.05. Assume that the process mean of the in-control process is not equal to the target value. By solving Eq. (8), we obtain the optimum solution with T * =6.9, * = 10.7, and ETC 1 (T *, * ) = 13.09. Next, assume that the process mean of the in-control process is equal to the target value. Then by solving Eq. (8), we obtain the optimum solution with T * =6.31andETC 1 (T *, * )= 1344.011. 6. Sensitivity Analysis Tables 1 list the sensitivity analyses of some parameters by considering the process mean that is either equal to or not equal to the target value for an incontrol process. From Tables 1, one has the follow consequences: 1. The demand quantity in units per year (W), the production rate in units per day (p), the demand quantity in units per day (d), the holding cost per unit item per year (B), and the set-up cost for each production run (S) have a major effect on the production run length and expected total cost per year for both the modified EMQ models.. The process standard deviation ( ) and the target value (m) have a major effect on the process mean for the modified EMQ model if the process mean is not equal to the target value for an in-control process. 3. For an in-control process, the optimum production run length and the expected total cost per year of the modified EMQ model if the process mean is not equal to the target value are less than those of the modified model if the process mean is equal to the target value. Hence, adopting the modified model of the latter will overestimate the expected total cost per year. 7. Discussion and Conclusion In this paper, the author presents a modified EMQ model based on the modified Chen and Chung s [31] model by considering that the process mean is not equal to the target value for an in-control process. Taguchi s [1] asymmetric quadratic quality loss function is applied in evaluating a product s quality. The numerical result shows that for an in-control process, the modified EMQ model when the process mean is not equal to the target value has smaller expected total cost per year than

16 Chung-Ho Chen Table 1. Sensitivity analyses of parameters for the modified EMQ model considering that the process mean is not equal to the target value for an in-control process W T * * ETC 1 (T *, * ) 16000 5.63 10.6 1176.51 18000 5.97 10.6 151.110 0000 6.9 10.7 13.09 000 6.60 10.7 1389.81 4000 6.89 10.8 1454.906 d T * * ETC 1 (T *, * ) 60 4.46 10.7 1844.044 70 5.15 10.7 1605.118 80 6.9 10.7 13.09 90 7.5 10.8 1153.534 100 1.440 10.8 695.314 T * * ETC 1 (T *, * ) 0.03 6.31 10.17 185.477 0.04 6.30 10.1 1301.471 0.05 6.9 10.7 13.09 0.06 6.8 10.33 1347.151 0.07 6.6 10.38 1376.83 p T * * ETC 1 (T *, * ) 85 13.500 10.7 749.75 95 7.44 10.7 118.178 100 6.9 10.8 13.09 110 4.91 10.7 153.768 10 4.07 10.7 1687.760 B T * * ETC 1 (T *, * ) 6 8.10 10.7 1038.713 8 7.03 10.7 1189.3 10 6.9 10.7 13.09 1 5.75 10.7 144.198 14 5.33 10.8 155.777 T * * ETC 1 (T *, * ) 0.3 6.3 10.7 1315.553 0.5 6.3 10.7 1316.694 1.0 6.9 10.7 13.09 1.5 6.5 10.7 1330.88.0 6.0 10.7 1343.19 S T * * ETC 1 (T *, * ) 16 5.63 10.7 1187.805 18 5.97 10.9 156.787 0 6.9 10.7 13.09 6.60 10.7 1384.079 4 6.89 10.8 1443.365 T * * ETC 1 (T *, * ) 0.03 6.30 10.7 1319.36 0.04 6.30 10.7 130.73 0.05 6.9 10.7 13.09 0.06 6.9 10.7 133.8 0.07 6.8 10.7 134.485 m T * * ETC 1 (T *, * ) 6 6.9 06.7 13.09 8 6.9 08.7 13.09 10 6.9 10.7 13.09 1 6.9 1.7 13.09 14 6.9 14.7 13.09 k 1 T * * ETC 1 (T *, * ) 1 6.9 10.7 13.09 1.5 6.8 10.7 1350.576 k T * * ETC 1 (T *, * ) 1.5 6.9 10.6 13.09.0 6.9 10.7 13.09.5 6.9 10.7 13.09 3.0 6.9 10.7 13.09 Table. Sensitivity analyses of parameters for the modified model considering that the process mean is equal to the target value for an in-control process W T * ETC 1 (T *, * ) 16000 5.64 1194.336 18000 5.98 171.019 0000 6.31 1344.011 000 6.61 1413.856 4000 6.90 1480.977 d T * ETC 1 (T *, * ) 60 4.47 1866.843 70 5.15 167.604 80 6.31 1344.011 85 7.8 1175.105 95 1.530 0714.875 T * ETC 1 (T *, * ) 0.03 6.3 193.388 0.04 6.31 1315.536 0.05 6.31 1344.011 0.06 6.30 1378.811 0.07 6.9 1419.935 k 1 T * ETC 1 (T *, * ) 1 6.31 1344.011 1.5 6.8 1361.571 k T * ETC 1 (T *, * ) 1.5 6.30 1333.01.0 6.31 1344.011.5 6.31 1354.999 3.0 6.3 1365.986 S T * ETC 1 (T *, * ) 16 5.63 1187.805 18 5.97 156.787 0 6.9 13.09 6.60 1384.079 4 6.89 1443.365 p T * ETC 1 (T *, * ) 85 13.590 0768.915 95 7.46 103.669 100 6.31 1344.011 110 4.91 1555.363 10 4.08 1710.738 B T * ETC 1 (T *, * ) 6 8.13 1059.934 8 7.05 110.899 10 6.31 1344.011 1 5.76 1464.416 14 5.33 1575.183 T * ETC 1 (T *, * ) 0.3 6.33 1339.141 0.5 6.33 1339.64 1.0 6.31 1344.011 1.5 6.7 135.508.0 6.1 1364.75 T * ETC 1 (T *, * ) 0.03 6.31 134.473 0.04 6.31 1343.58 0.05 6.31 1344.011 0.06 6.30 1344.733 0.07 6.30 1345.46

Optimum Production Run Length and Process Mean Setting 163 that of the modified EMQ model when the process mean is equal to the target value. One should consider using this modified EMQ model in order to reduce overestimation of the expected total cost per year. Further study should address the problem of the larger-is-better, smaller-is-better, or multivariable quality loss function applied in formulating the modified EMQ model. References [1] Porteus, E. L., Optimal Lot Sizing, Process Quality Improvement and Set-Up Cost Reduction. Operations Research, Vol. 34, pp. 137 144 (1986). [] Rosenblatt, M. J. and Lee, H. L., Economic Production Cycles with Imperfect Production Processes, IIE Transactions, Vol. 17, pp. 48 54 (1986a). [3] Rosenblatt, M. J. and Lee, H. L., A Comparative Study of Continuous and Periodic Inspection Policies in Deteriorating Production Systems, IIE Transactions, Vol. 18, pp. 9 (1986b). [4] Lee, H. L. and Rosenblatt, M. J., Simultaneous Determination of Production Cycle and Inspection Schedules in a Production System, Management Science, Vol. 33, pp. 115 1136 (1987). [5] Lee, H. L. and Rosenblatt, M. J., A Production and Maintenance Planning Model with Restoration Cost Dependent on Detection Delay, IIE Transactions, Vol. 1, pp. 368 375 (1989). [6] Lee, J. S. and Park, K. S., Joint Determination of Production Cycle and Inspection Intervals in a Deteriorating Production System, Journal of the Operational Research Society, Vol. 4, pp. 775 783 (1991). [7] Rahim, M. A., Joint Determination of Production Quantity, Inspection Schedule, and Control Chart Design, IIE Transactions, Vol. 6, pp. 11 (1994). [8] Rahim, M. A. and Ben-Daya, M., A Generalized Economic Model for Joint Determination of Production Run, Inspection Schedule and Control Chart Design, International Journal of Production Research, Vol. 36, pp. 77 89 (1998). [9] Al-Sultan, K. S., Introduction to Optimization, In: Optimization in Quality Control, Al-Sultan, K. S. and Rahim, M. A., eds, Kluwer Academic Publishers, Boston, pp. 3 53 (1997). [10] Wright, C. M. and Mehrez, A., An Overview of Representative Research of the Relationships between Quality and Inventory, Omega, Vol. 6, pp. 9 47 (1998). [11] Tahera, K., Chan, W. M. and Ibrahim, R. N., Joint Determination of Process Mean and Production Run: A Review, International Journal of Advanced Manufacturing Technology, Vol. 39, pp. 388 400 (008). [1] Sana, S. S., Goyal, S. K. and Chaudhuri, K., On a Volume Flexible Inventory Model for Items with an Imperfect Production System, International Journal of Production Research, Vol., pp. 64 80 (007a). [13] Sana, S. S., Goyal, S. K. and Chaudhuri, K., An Imperfect Production Process in a Volume Flexible Inventory Model, International Journal of Production Economics, Vol. 105, pp. 548 559 (007b). [14] Sana, S. S., A Production-Inventory Model in an Imperfect Production Process, European Journal of Operational Research, Article in press (009a). [15] Sana, S. S., An Economic Production Lot Size Model in an Imperfect Production System, European Journal of Operational Research, Article in press (009b). [16] Jang, J. S., Ahn, D. G., Lee, M. K. and Elsayed, E. A., Optimum Initial Process Mean and Production Cycle for Processes with a Linear Trend, Quality Engineering, Vol. 13, pp. 9 35 (000). [17] Young, J. K., Cho, B. R. and Phillips, M. D., Determination of the Optimal Process Mean with the Consideration of Variance Reduction and Process Capability, Quality Engineering, Vol. 13, pp. 51 60 (000). [18] Lee, M. K. and Elsayed, E. A., Process Mean and Screening Limits for Filling Processes under Two- Stage Screening Procedure, European Journal of Operational Research, Vol. 138, pp. 118 16 (00). [19] Anis, M. Z., Determination of the Best Mean Fill, Quality Engineering, Vol. 15, pp. 407 409 (003). [0] Hong, S. H. and Cho, B. R., Joint Optimization of Process Target Mean and Tolerance Limits with Measurement Errors under Multi-Decision Alternatives, European Journal of Operational Research, Vol. 183, pp. 37 335 (007). [1] Taguchi, G., Introduction to Quality Engineering, Tokyo: Asian Productivity Organization (1986). [] Kim, Y. J. and Cho, B. R., Determining the Optimum Process Mean for a Skewed Process, International Journal of Industrial Engineering Theory Applications and Practice, Vol. 10, pp. 555 561 (003).

164 Chung-Ho Chen [3] Rahim, M. A. and Tuffaha, F., Integrated Model for Determining the Optimal Initial Settings of the Process Mean and the Optimal Production Run Assuming Quadratic Loss Functions, International Journal of Production Research, Vol. 4, pp. 381 3300 (004). [4] Chan, W. M. and Ibrahim, R. N., Evaluating the Quality Level of a Product with Multiple Quality Characteristics, International Journal of Advanced Manufacturing Technology, Vol. 4, pp. 738 74 (004). [5] Chan, W. M., Ibrahim, R. N. and Lochert, P. B., Quality Evaluation Model Using Loss Function for Multiple S-Type Quality Characteristics, International Journal of Advanced Manufacturing Technology, Vol. 6, pp. 98 101 (005a). [6] Chan, W. M., Ibrahim, R. N., and Lochert, P. B., Evaluating the Product Quality Level under Multiple L-Type Quality Characteristics, International Journal of Advanced Manufacturing Technology, Vol. 7, pp. 90 95 (005b). [7] Teeravaraprug, J., Determining Optimal Process Mean of Two-Market Products, International Journal of Advanced Manufacturing Technology, Vol. 5, pp. 148 153 (005). [8] Chen, C. H., The Optimum Selection of Imperfect Quality Economic Manufacturing Quantity and Process Mean by Considering Quadratic Quality Loss Function, Journal of the Chinese Institute of Industrial Engineers, Vol. 3, pp. 1 19 (006a). [9] Chen, C. H., The Modified Pulak and Al-Sultan s Model for Determining the Optimum Process Parameters, Communications in Statistics Theory and Methods, Vol. 35, pp. 1767 1778 (006b). [30] Pulak, M. F. S. and Al-Sultan, K. S., The Optimum Targeting for a Single Filling Operation with Rectifying Inspection, Omega, Vol. 4, pp. 77 733 (1996). [31] Chen, S.-L. and Chung, K.-J., Determining of the Optimal Production Run and the Most Profitable Process Mean for a Production Process, International Journal of Production Research, Vol. 34, pp. 051 058 (1996). Manuscript Received: Feb. 0, 009 Accepted: Jul. 3, 009